EARTHQUAKE RESISTANT TENSILE LAP SPLICES

Downloaded from ascelibrary.org by New York University on 05/11/15. Copyright ASCE. For personal use only; all rights reserved.

By J. Dario Aristizabal-Ochoa, 1 M . ASCE

ABSTRACT: A review of current code requirements on lap splices and their lim-
itations in seismic design is presented. A simple analytical model of an embed-
ded bar is examined and correlated with available experimental data. The force
transfer mechanisms of lap splices and the effects of properly placed transverse
reinforcement to avoid brittle failure are studied. The author concluded that:
(1) Concentrated transverse reinforcement at the ends of lap splices and in con-
tact with the longitudinal reinforcement is structurally beneficial; and (2) sharp
offset bends must be avoided to prevent fatigue failure in the lapped reinforce-
ment and to prevent excessive bursting forces in concrete in the event of severe
tensile load reversals.

INTRODUCTION

The specific problem considered in this paper is design criteria for ten-
sion lap splices in regions of potentially severe stress reversals. This sit-
uation occurs in critical members, such as structural walls and columns,
where splices of tensile reinforcement bars are unavoidable. During se-
vere earthquakes, alternate overturning forces can induce inelastic stress
reversals in the lapped bars. Bond strength of the concrete and pene-
tration of yielding into the development length must be controlled in
order to guarantee the effectiveness, strength, and stiffness of lap splices.
At present, the designer has no guidance on effectiveness of tensile lap
splices in critical members.
The main advantages of lap splices over welded and mechanically con-
nected splices are relative simplicity and low cost of installation in the
field. However, given insufficient experimental data on lap splices, most
building codes have penalized their use.
This paper reviews current code requirements on lap splices and their
limitations in seismic design, particularly when they apply to tension
members, such as boundary elements of structural walls and exterior
columns under severe tensile forces. A literature review on lap splices
is presented as well as a summary of the researcher's experimental work
on tension lap splices (1,2). A simple analytical model of an embedded
bar is examined and correlated with available experimental results. The
force transfer mechanism in lap splices and the effects of properly placed
transverse reinforcement to avoid bond failure are reviewed. Rational
recommendations for the seismic design of lap splices are given.
lAsst. Prof, of Civ. Engrg., Vanderbilt Univ., Nashville, Tenn. 37235.
Note.—Discussion open until September 1, 1983. To extend the closing date
one month, a written request must be filed with the ASCE Manager of Technical
and Professional Publications. The manuscript for this paper was submitted for
review and possible publication on February 19,, 1982. This paper is part of the
Journal of Structural Engineering, Vol. 109, No. 4, April, 1983. ©ASCE, ISSN
0733-9445/83/0004-0843/$01.00. Paper No. 17896.
843

J. Struct. Eng. 1983.109:843-858.

 CURRENT BUILDING PROVISIONS

The 1977 American Concrete Institute (ACI) Building Code (8) and the
1976 Uniform Building Code (32) require lap lengths greater than the
development length and no lap splices are permitted for bars larger than
No. 11 (4> = 35 mm). Table 1 summarizes the 1977 ACI Building Code
Downloaded from ascelibrary.org by New York University on 05/11/15. Copyright ASCE. For personal use only; all rights reserved.

provisions for tension lap splices in flexural members. Three classes of

lap splices (A, B, and C) are defined, depending on the ratio of the steel
provided to the steel required and the percentage of the bars inter-
rupted. For instance, when 75% or less of the bars at the section in ques-
tion are interrupted and the tensile steel provided is equal to or greater
than two times that required, a Class A splice is required. This splice
has a lap length equal to the bar development length, td. Seismic design
provisions in Appendix A of the 1977 ACI Building Code (8) also require
that lap lengths be at least 24 nominal bar diam or 12 in. (305 mm).
The basic intent of the building codes is to discourage use of tensile
splices in regions of critical stress and to encourage staggering of splices.
For seismic conditions, applicability of current codes provisions for ten-
sile lapped splices needed to be clarified. The following questions arise:

1. Lap lengths are usually based on development of 125% of the yield

stress of the bar. Is this appropriate for seismic conditions when severe
load reversals can quickly deteriorate the bond strength of the concrete
(6,7,17)?
2. Experimental investigations (21) indicate a significant improvement
in performance with ordinary transverse reinforcement. For seismic de-
sign, unusually large amounts of transverse reinforcement are used. What
are the effects of such reinforcement on performance of lap splices under
load reversals?
3. Both the amount and the distribution of transverse reinforcement
have a critical effect on ductility, strength, and behavior of lap splices
(2). What is the optimum arrangement of transverse reinforcement to
control bond deterioration in the concrete and give maximum strength
and ductility for a given lap splice configuration?
TABLE 1.—Tension Lap Splices According to 1977 ACI Building Code
Maximum Percent of As Spliced
within Required Lap Length
As provided/A, required 50 75 100
(1) (2) 0) (4)
Equal to or greater than 2 Class A Class A Class B
Less than 2 Class B Class C Class C
Note: A, provided/A, required = ratio of area of reinforcement provided to area
of reinforcement required by analysis of splice location; Class A splice = 1.0^;
Class B splice = 1.3€d; Class C splice = 1.7€d; td = development length, in inches
= 0.04 Aify/Vf,., but not less than 0.0004 dhfy for No. 11 bars or smaller in normal
weight concrete; fy = specified yield strength of reinforcement, in pounds per
square inch; db = nominal diameter of bar, in inches; Ab = area of individual bar,
in square inches; and fc = specified compressive strength of concrete, in pounds
per square inch. 1 in. = 25.4 mm; 1 sq in. = 645 mm2; and 1 psi = 6.89 kPa.

844

J. Struct. Eng. 1983.109:843-858.

 lACKOROUND

Literature Review.—A review of literature on lap splices of reinforcing

bars indicates that very little work has been done to determine perfor-
mance under seismic loading conditions. Appendix I is a list of refer-
ences covering the topic of lap splices.
An evaluation of development length and lap splices has been re-
Downloaded from ascelibrary.org by New York University on 05/11/15. Copyright ASCE. For personal use only; all rights reserved.

ported by Orangun, et al. (21). This paper describes the derivation of

an equation for development length that includes effects of concrete cov-
ers, bar spacings, and small percentages of uniformly distributed trans-
verse reinforcement. However, their recommendations are based on tests
of monotonically loaded members. These recommendations must be
reevaluated for members subjected to seismic loading and designed with
large percentage of transverse reinforcement.
Tests of lap splices under monotonic compressive loads have been re-
ported by Arthur and Cairns (3). They concluded that:

1. Forces in compression lap splices are transferred by bond stresses

around the circumference of the bars and by end-bearing of the bars on
the concrete.
2. Transfer of forces causes bursting stresses on the surrounding
concrete.
3. Strength in bond and end-bearing is dependent on the resistance
available to counteract bursting stresses.
4. Transverse reinforcement located at ends of the splice is more ef-
fective in resisting bursting stresses.

Arthur and Cairns developed an expression to predict stresses in the

main steel. Their expression includes the contributions of transverse re-
inforcement, bond, and end bearing. They also recommended that one-
third of required transverse reinforcement be placed within a distance
of 15% of the lap length from each end of the splice.
Tests of tension lap splices in beams have recently been reported by
Betzle (4). His tests included beams longitudinally reinforced with No.
5 and No. 9 bars without transverse reinforcement. Specimens were sub-
jected to monotonic loading. Betzle concluded that bond stresses in the
concrete and force transfer between lapped bars are mainly concentrated
at the ends of the splice.
Tests of bond and anchorage of reinforcing bars under inelastic re-
versals of load.have been reported by Ismail and Jirsa (18), Hassan and
Hawkins (14), and Popov (25). These tests indicate that rate of bond
deterioration and response of anchored bars are significantly affected by
loading history. Slip and deformation of anchored bars under load re-
versals were observed to have substantial influence on the hysteretic
response of the members tested. This was also observed in cantilever
beam tests by Brown and Jirsa (7). They concluded, "Deformation of the
steel in the anchorage zone (within the fixed end) contributed signifi-
cantly to the total deformation and to the energy absorbing capacity of
the specimens."
Thus, for sesimic conditions, the load vs. slip relationship of bars is
of prime importance. Brown and Jirsa (7), and Hawkins (15) also con-
845

J. Struct. Eng. 1983.109:843-858.

 TABLE 2.—Test Specimens (After Ref. 2)
Load Transverse Longitudinal
Specimen" history reinforcementb reinforcementb
0) (2) (3) (4)
S6-1 monotonic No. 3 @ 2" (51 mm) 8 No. 6
S6-2 reversing No. 3 @ 2" 8 No. 6
Downloaded from ascelibrary.org by New York University on 05/11/15. Copyright ASCE. For personal use only; all rights reserved.

S6-3 reversing No. 3 @ 4" (102 mm) 8 No. 6

S6-4 monotonic No. 3 @ 4" 8 No. 6
S8-1 monotonic No. 3 @ 2" 4 No. 8
S8-2 reversing No. 3 @ 2" 4 No. 8
S8-3 monotonic No. 3 @ 12" (305 mm) 4 No. 8
S8-4 reversing No. 3 @ 4" 4 No. 8
"AH specimens 3,000 psi (20.7 MPa) design compressive strength concrete.
b
Bars conforming to ASTM designation A615 grade 60. Longitudinal reinforce-
ment was spliced at the same location.

elude that for cyclic inelastic loads, it may be necessary to increase

embedment lengths required by the 1977 ACI Building Code (8).
Researcher's Experimental Work.—An experimental program which
evaluated the performance of Class C splices under severe load reversals
has been reported by the writer (1,2). A series of eight test specimens
containing No. 8 and No. 6 bars was tested. Variables in this experi-
mental program (Table 2) included the load history, the configuration of
the lapped reinforcement, and the amount of transverse reinforcement.
In each specimen, 100% of the longitudinal reinforcement was spliced
at the same location.
The tests were performed on I-shaped specimens, as shown in Fig. 1.
Axial loads were applied through the two end blocks using two 100-ton
(980-kN) hydraulic actuators. The splice test region was at the center of
the column portion of the specimen. The column portion had cross-sec-
tional dimensions of 12 in. x 12 in. (305 mm x 305 mm) and a length
of 96 in. (2,440 mm).
Following common practice, corner bars had offsets at the start of the
lap. The slope of the offset was 1:6, the maximum permitted by the ACI
Code (8). Photographs of the reinforcement at the location of the offset
are shown in Fig. 2. The interior bars of specimens with No. 6 bars were
not offset.

#^IED ( .£f^ #
LAP SPLICE
(1.42 m)

00 #
(1.22 m) (2.44 m) (1.22 m)

FIG. 1.—Test Specimen Dimensions

846

J. Struct. Eng. 1983.109:843-858.

Downloaded from ascelibrary.org by New York University on 05/11/15. Copyright ASCE. For personal use only; all rights reserved.

FIG. 2.—Reinforcement Details at the Offset of: (a) Test Specimen §8-2 and; (b)
Test Specimen S6-2

The amount of transverse reinforcement varied from that required for

seismic design conditions according to the Union Building Code (UBC)
Code (32) to maximum column tie spacing according to the ACI Code
(8), as shown in Fig. 3. The amount of transverse reinforcement in each
test specimen is listed in Table 2. Four tests under tensile monotonic
loading and four tets under severe reversals of loading were carried out.
A typical history of load reversals is shown in Fig. 6. Complete details
of these experiments are given elsewhere (1,2). The main observations
and conclusions based on the eight tests are the following:

1. Specimens designed as Class C lap splices with transverse rein-

forcement for seismic conditions according to the 1976 UBC Code (32)
performed well under monotonic and reversing loads. The strength of

FIG. 3.—Lap Splice of: (a) Test Specimen S8-2; (ft) Test Specimen 88-3
847

J. Struct. Eng. 1983.109:843-858.

 the specimens varied from 169%-174% of the design yield load.
2. Strength of the specimens was not affected significantly by load
history.
3. All specimens experienced large post-yield elongations. Specimens
subjected to monotonic loading exhibited slightly larger ductilities than
those subjected to load reversals.
Downloaded from ascelibrary.org by New York University on 05/11/15. Copyright ASCE. For personal use only; all rights reserved.

4. The use of offset bars at the end of the lap caused serious local
distress. The extent of damage was larger in specimens with large bars.
Moreover, this type of reinforcing detail may cause low cycle fatigue
failure in the reinforcement in structures under reversing loads.
5. In an eight-bar arrangement (four corner bars and four interior bars,
Fig. 2(b), with 100% of the bars spliced, slip of interior bars controlled
capacity. Splitting cracks first appeared along interior bars at the offset,
and then propagated from both end regions into the splice. The cracks
were first observed as the load approached yield. As the load ap-
proached ultimate, interior spliced bars slipped and transferred load to
the corner bars. Longitudinal splitting was not as well contained for in-
terior bars as it was for corner bars.
6. Transverse reinforcement was effective in controlling longitudinal
splitting and bar slip as well as yield penetration along the sliced bars.
7. The amount and distribution of transverse reinforcement has a crit-
ical effect on ductility, strength, and behavior of lap splices. An insuf-
ficient amount of hoop reinforcement at the ends of a splice can lead to
reduction in ductility and strength, and to severe damage within the
splice region. From measurements of strains in transverse reinforce-
ment, it appears that closely spaced hoops at the ends of a splice would
improve performance. Hoops at the ends of the splices were more ef-
fective than interior hoops in resisting splitting and bursting of the
concrete.
8. The 1976 UBC Code requirements (32) for lateral reinforcement con-
finement in seismic regions are rather conservative. In specimens with
four corner bars (S8 Series in Table 2), a reduction of 83% in the required
transverse reinforcement caused a reduction of only 15% in the mea-
sured ultimate strength. In specimens with an eight-bar arrangement (S6
Series in Table 2), a reduction of 50% in the required transverse rein-
forcement caused a reduction of 15% in the measured ultimate strength.
9. Transverse hoops must be in contact with longitudinal bars to be
effective. This is based on experimental data taken from the hoops at
the offset which were not in contact with the longitudinal bars.

FORCE TRANSFER MECHANISMS IN TENSILE LAP SPLICES

Description of the Mechanisms.—Forces between lapped bars in ten-

sion are transferred from one bar to the other through the concrete by
bond stresses around the circumference of the bars (5,9,10,11,12,16,24,26,
27,28,29,30,31,33). The bond stresses induce radial or "bursting" forces
in the concrete surrounding the splice (13,19,20,23). Such radial forces
come largely from wedging action when the ribs of the bar bear against
the concrete. These forces are heavily concentrated at the ends of the
splice (1,2). When these bursting forces are not properly controlled, bond
failure ensues, resulting in splitting of the concrete along the lapped bars
848

J. Struct. Eng. 1983.109:843-858.

 Non-offset "outside" Offset
Downloaded from ascelibrary.org by New York University on 05/11/15. Copyright ASCE. For personal use only; all rights reserved.

FIG. 4.—(a) Lap Splice Details; (b) Internal Couple Created by the Bar Offset

and subsequent pull-out of the bars.

Along the lap splice, the bursting or radial forces are resisted by the
concrete cover, the transverse reinforcement, and the dowel action of
the "outside" bars. However, at the offset, radial forces are resisted mainly
by the transverse reinforcement. The contribution of concrete cover at
the offset end is reduced by the additional bursting forces induced by
the internal couple because of the offsetting of the bars (Fig. 4). This
explains the high concentration of damage in the offset region of lap
splices.
Figures 5, 6, and 7 show typical distributions of crack widths and strains
in longitudinal and transverse reinforcement of a lap splice (1,2). Figure
5 shows the growth of the concrete cracks with load reversals, partic-
ularly at the offset end of the splice. Figure 6(b) shows the antisymmetry
in the strain along lapped bars, the penetration of yielding, and the ten-
dency of the strain distribution to become linear as applied load ex-
ceeded the yield level of the bars. Figure 7 shows strain in the transverse
reinforcement at different locations. Strains in the steel hoops 1, 2, 3,
and 6 indicate that most of the induced bursting forces are concentrated
at the end of the splice, particularly at the offset end. These results in-
dicate that most of the forces in the bars are transferred at the end of
the splice and that a better performance could be obtained by redistri-
buting the hoops. Concentrated hoops at the ends, particularly at the
offset end, would be structurally beneficial.
Strains in concrete and reinforcement are more heavily concentrated
at the offset end of the splice than at the nonoffset end. The results
because the offset tends to "straighten out" when the longitudinal re-
inforcement is in tension. Therefore, an outward component of thrust
occurs at the offset, which generally causes spalling of the concrete cover.
As a consequence, the contribution of end bearing of the bars to the
force transfer when the lapped reinforcement is subjected to compres-
sion forces is practically reduced to zero.
Main Design Parameters.—The severity and extent of concrete burst-
ing depend on: (1) The relative size of the offset bars with respect to the
size of the cross section (i.e., the larger the internal couple created by
the lapped bars, the more severe is the extent of concrete bursting); (2)
the amount and configuration of lapped reinforcement; and (3) amount
and distribution of transverse reinforcement.
849

J. Struct. Eng. 1983.109:843-858.

 / VCycle 4
Crack OlOO ,' \ Crack
Width, Widlh,
\—— Cycl e3
CycleE
*nM
V. \ ~//y\l
Downloaded from ascelibrary.org by New York University on 05/11/15. Copyright ASCE. For personal use only; all rights reserved.

1.0 Tensile lood= 242 kips (1076 kN)

at cycles I, 2, 3, and 4

FIG. 5.—Crack Widths in the Concrete along the Lap Splice of Test Specimen S6-
2 (After Ftef. 2)

a)
-Loading History-

6 Cycle No.

Level of Compreitlve Load 0,72 R,

I in. = 25.4 mm
I M.p = 4 . 4 4 6 kN

- Yield Level
~P= 337 K.OJ / T\

b) («) y/i
0,001 Strain

i i i l i r i

Slrain o,OOi 0.001 Strain

/ r~~</'v-\
-'" iC-'' !_![__
/ ' / p s 337 kips (M
0 002

- Yield Level

Note = Number within parenthesis irtdlcatna eyele

M ' monotonia loading after sixth cycle

FIG. 6.—(a) Reversing Load History; (b) Strains In the Longitudinal Reinforcement
(After Ret. 2)
850

J. Struct. Eng. 1983.109:843-858.

 Load,
Downloaded from ascelibrary.org by New York University on 05/11/15. Copyright ASCE. For personal use only; all rights reserved.

°0 0.001 0.002 I 0.003

Strain

PIG. 7.—Tensile Strains In the Transverse Reinforcement of Test Specimen S6-2

(After flef. 2)

Propagation of concrete bursting and yield penetration from the ends

of the splice into the splice depends mainly on the amount and distri-
bution of transverse reinforcement. Bursting forces in the concrete pro-
duce splitting cracks along the longitudinal lapped reinforcement, and
subsequent loss of bond and yield penetration, both of which reduce
the effective anchorage length of the bars. Such a failure mechanism is
mainly resisted by the transverse reinforcement. Although some initial
resistance is also obtained from the splitting strength of the concrete
cover, this must be considered negligible for most practical purposes.
Lap splices with light or no transverse reinforcement have low ductility
and generally fail violently by concrete bond failure or bar pull-out at
loads lower than that corresonding to the strength of the bars. In con-
trast, lap splices heavily reinforced with uniform transverse reinforce-
ment fail at the offset by bar fracture.
The amount and configuration of lapped reinforcement are important
factors in lap splice performance under severe load reversals. The larger
the lapped bars relative to the size of the cross section, the larger are
the internal couples created by the offset of the bars and consequently
the larger the bursting forces in the concrete, particularly, at the offset
end of the splice. When corner and interior bars are spliced all together
as shown in Figs. 2(b) and 8(a), slip of the interior bars generally controls
the splice tensile capacity. Splitting cracks start along the interior bars
at the offset end and propagate from both ends into the splice. As the
load is increased, the interior bars slip and transfer the applied load to
the corner bars (1,2). To avoid this type of mechanism and control split-
ting, interior spliced bars must also be contained by supplementary
transverse reinforcement in contact with the longitudinal bars (Fig. 8(b)).

EMBEDDED BAR MODEL

This section presents a simple linear elastic model of a bar embedded

851

J. Struct. Eng. 1983.109:843-858.

 SPLICED
BARS

po. :•..• :cc

TRANSVERSE f "HkP'.'.QQ • - ' - Q J '

1EINFORCEMENT ".yT . . . r - ? l
Downloaded from ascelibrary.org by New York University on 05/11/15. Copyright ASCE. For personal use only; all rights reserved.

n SUPPLEMENTARY
TRANSVERSE
REINFORCEMENT

b)

FIG. 8.—Typical Cross-Section Configurations of Lap Splices (a) Single "Hoop

Reinforcement; (b) Additional Supplementary Reinforcement

o
• '. • ' ' ' , ' & •

\% 1
1

V * •

'^&-:-
•a dx

AVERAGE CONCRETE
STRESS

FIG, 9.—Model of an Embedded Bar in Concrete

852

J. Struct. Eng. 1983.109:843-858.

 in concrete and subjected to tensile load P0 at the free end. Figure 9
shows the mechanical model and the equilibrium of a bar element and
section x from the embedded end of the bar. Assuming that the bonding
force per unit length, t, is proportional to the relative deformation be-
tween the bar and the surrounding concrete and also assuming average
stresses and strains, the following equations can be obtained:
Downloaded from ascelibrary.org by New York University on 05/11/15. Copyright ASCE. For personal use only; all rights reserved.

dP
Tx = t • ;- (lfl)

t = k(Ub-Uc) (lb)
N = P0-.P (2)
in which P = tensile force in the embedded bar at section x; t = average
bond force per unit length of bar at section x; k = constant; Ub = average
elongation of bar at section x; and Uc = average deformation of concrete
surrounding at the bar at section x. Differentiating Eq. la-b and assum-
ing strain proportional to force, we obtain
d2P dt
-^ = — = *(e» - ec) = k(ktP -k2N) (3)
in which ej, = average tensile strain in the bar (= dUb/dx = k^P); ec =
average tensile strain in the concrete (= dUJdx = k2N); N = total force
in the concrete block at section x; kx = constant; and k2 = constant. Rear-
ranging Eq. 3 and substituting in Eq. 2, the following differential equa-
tion is obtained:

BOND STRESS

FIG. 10.—Distribution of Bar Forces and Concrete Bond Stresses along an

Embedded Bar

853

J. Struct. Eng. 1983.109:843-858.

Downloaded from ascelibrary.org by New York University on 05/11/15. Copyright ASCE. For personal use only; all rights reserved.

YIELD PENETRATION

BOND STRESS
NOTES:
a-Just before bar yield
b-After some bar yield
c-Just before ultimate load

FIG. 11.—Typical Distributions of Bar Forces and Concrete Bond Stresses at Dif-
ferent Loading Levels

CONFINED CONCRETE AT LEAST .151

FROM FREE END

BOND STRESS
NOTES:
a-Just before bar yield
b-After some bar yield
c-Just before ultimate load

FIG. 12.—Expected Distribution of Concrete Bond Stresses when Concrete

Confined at the Beginning of the Embedment

854

J. Struct. Eng. 1983.109:843-858.

 d2P .
—I - a2P = fP0 . , (4)
dx
in which a2 = k(ki + k2)> and j32 = kk2. Applying the boundary conditions
shown in Fig. 9, the solutions for P and t are as follows:
2
P\ 2 P0
Downloaded from ascelibrary.org by New York University on 05/11/15. Copyright ASCE. For personal use only; all rights reserved.

P= Sinh ax + Sinh a€ - Sinh a(€ - x) h

a/ Sinh a€
2 _ 2

. . - £a/ Sinha€ (.LVP

Cosh ax + Cosh a{£ - x)\ (5)

Figures 10(c) and 10(b) show the distribution of P and t along the bar.
The distribution of P along the bar shown in Fig. 10(a) is similar to that
obtained experimentally and shown in Fig. 6(b) before yielding of the
embedded bar. A similar distribution was obtained experimentally by
Orr (22).
Although the model presented is simple and applies only in the elastic
range, it explains the concentration of concrete bond stresses at the ends
of the embedded bar and the stress distribution along the bar. It also
indicates that the inelastic behavior in concrete and steel starts at the
beginning of the embedment.
Two basic modes of failure can be expected in an embedded bar under
tension: failure in the bar itself with the full strength of the bar being
developed (generally termed as "bar fracture"), and failure in the con-
crete by bonding (generally termed as "bar pullout"). Figure 11 shows
typical distributions of bar stresses and concrete bond stresses at differ-
ent load levels of an embedded bar under tension. Curves b and c in the
figure show the continuous reduction of the effective embedment length
and yield penetration with load. If the area under the bond stress curve
is less than the applied force, P0, "bar pullout" is the mode of failure.
Thus, in order to guarantee 1.25 times the yield bar strength, most build-
ing codes require an embedment length much larger than the devel-
opment length. However, the addition of lateral or hoop reinforcement
increases the bonding strength of the surrounding concrete and also
controls bond deterioration. Figure 12 shows the expected distribution
of concrete bonding stress in an embedded bar with hoop reinforcement.
The net effect of hoop reinforcement is the reduction of required embed-
ment length.
This model and the experimental results cited previously (1,2) indicate
that hoop reinforcement is necessary and very beneficial in the region
close to the beginning of the embedment. It is recommended that this
region should be taken as 15% of the development length (3).

CONCLUSIONS

The following conclusions and recommendations can be drawn:

1. Transverse reinforcement has significant beneficial effects on the

strength and ductility of lap splices.
855

J. Struct. Eng. 1983.109:843-858.

 2. The use of concentrated transverse hoops at the ends of lap splices
and in contact with the longitudinal reinforcement and supplementary
transverse hoops for interior lapped bars can prevent bond failure and
can fully develop the ultimate strength and ductility of the lapped bars.
3. Transverse hoops concentrated at both ends of lap splices within a
distance of 1/6 of the required splice length are recommended in seismic
Downloaded from ascelibrary.org by New York University on 05/11/15. Copyright ASCE. For personal use only; all rights reserved.

design. The concentrated transverse hoops shall be placed at the mini-

mum acceptable spacing and in contact with the lapped bars. The other
2/3 of the splice length shall have minimum hoop reinforcement to avoid
bar bucking or any other instability consideration.
4. Local bending or offset of the lapped bars should be avoided in
seismic design, particularly in bars greater than No. 6 bar, or, at least,
the offset should be made as smooth as possible. A maximum slope of
1/10 is recommended for bars No. 8 and greater. The transverse hoops
supporting the offset bends should be in contact with the longitudinal
reinforcement.
5. Research is recommended to determine the trade-offs between min-
imum lap splice length and provided transverse reinforcement. Test re-
sults indicate that: (1) ACI (Standard 318-77) provisions on lap splice
lengths in members with confinement reinforcement are overconserva-
tive; (2) splice lengths can be significantly reduced by the beneficial ef-
fects of the confinement steel.

APPENDIX.—REFERENCES

1. Aristizabal-Ochoa, J. D., Fiorato, A. E., and Corley, W. G., "Tension Lap

Splices Under Severe Load Reversals," Proceedings of the World Conference on
Earthquake Engineering, Sept., 1980, Vol. 7, pp. 55-62.
2. Aristizabal-Ochoa, J. D., Fiorato, A. E., and Corley, W. G., "Earthquake Re-
sistant Structural Walls—Tests of Lap Splices," Report to National Science
Foundation submitted by PCA Construction Technology Laboratories, Sept.,
1979.
3. Arthur, P. D., and Cairns, J. W., "Compression Laps of Reinforcement in
Concrete Columns," The Structural Engineer, No. 1, Vol. 56B, Mar., 1978, pp.
9-12.
4. Betzle, M., "Bond Slip and Strength of Lapped Bar Splices," Douglas McHenry
International Symposium on Concrete and Concrete Structures, American
Concrete Institute, SP 55-20, 1978, pp. 493-514.
5. "Bond Stress—The State of the Art," ACI Committee 408, Journal of the Amer-
ican Concrete Institute, Vol. 63, No. 11, Nov., 1966, pp. 1161-1189.
6. Bresler, B., and Bertero, V. V., "Behavior of Reinforced Concrete Under Re-
peated Load," Journal of the Structural Division, ASCE, Vol. 94, No. ST6, June,
1968, pp. 1567-1590.
7. Brown, R. H., and Jirsa, J. O., "Reinforced Concrete Beams Under Load Re-
versals," Journal of the American Concrete Institute, Vol. 68, No. 5, May, 1971,
pp. 380-390.
8. "Building Code Requirements for Reinforced Concrete (ACI 318-77)," Amer-
ican Concrete Institute, Detroit, Mich., 1977, p. 102.
9. Chinn, J., Ferguson, P. M., and Thompson, J. N., "Lapped Splices in Rein-
forced Concrete Beams," Journal of the American Concrete Institute, Vol. 52, No.
2, Oct., 1955, pp. 201-213.
10. Colaco, J. P., and Siess, C. P., "Behavior of Splices in Beam-Column Con-
nections," Journal of the Structural Division, ASCE, Vol. 93, No. ST5, Oct.,
1967, pp. 175-193.
856

J. Struct. Eng. 1983.109:843-858.

 11. Ferguson, P. M., "Small Bar Spacing or Cover—A Bond Problem for the De-
signer," Journal of the American Concrete Institute, Vol. 74, No. 9, Sept., 1977,
pp. 435-439.
12. Ferguson, P. M , and Breen, J. E., "Lapped Splices for High Strength Rein-
forcing Bars," Journal of the American Concrete Institute, Vol. 62, No. 9, Sept.,
1965, pp. 1063-1077.
13. Goto, Y., "Cracks Formed in Concrete Around Deformed Tension Bars," Journal
Downloaded from ascelibrary.org by New York University on 05/11/15. Copyright ASCE. For personal use only; all rights reserved.

of the American Concrete Institute, Vol. 68, No. 4, Apr., 1971, pp. 244-251.
14. Hassan, F. M., and Hawkins, N. M., "Effects of Post-Yield Loading Re-
versals on Bond Between Reinforcing Bars and Concrete," Report SM 73-2,
Department of Civil Engineering, University of Washington, Mar., 1973, 244
pp.
15. Hawkins, N. M., "Development Length Requirements for Reinforcement Bars
Under Seismic Conditions," Proceedings, Workshop on Earthquake-Resistant
Reinforced Concrete Building Construction, July, 1977, pp, 696-704.
16. "Interaction Between Steel and Concrete," ACI Committee 408, Journal of the
American Concrete Institute, Vols. 76 and 77, Nos. 1 and 2, Jan.-Feb., 1979.
17. Ismail, M. A. F., and Jirsa, J. O., "Bond Deterioration in Reinforced Concrete
Subject to Low Cycle Loads," Journal of the American Concrete Institute, Vol.
69, No. 6, June, 1972, pp. 334-343.
18. Ismail, M. A. F., and Jirsa, J. O., "Behavior of Anchored Bars Under Low
Cycle Overloads Producing Inelastic Strains," Journal of the American Concrete
Institute, Vol. 69, No. 7, July, 1971, pp. 433-438.
19. Lutz, L. A., and Gergely, P., "Mechanics of Bond and Slip of Deformed Bars
in Concrete," Journal of the American Concrete Institute, Vol. 64, No. 11, Nov.,
1967, pp. 711-721.
20. Nilson, A. H., "Internal Measurement of Bond Slip," Journal of the American
Concrete Institute, Vol. 69, No. 7, July, 1972, pp. 439-441.
21. Orangun, C. O., Jirsa, J. O., and Breen, J. E., "A Reevaluation of Test Data
on Development Length and Splices," Journal of the American Concrete Insti-
tute, Vol. 74, No. 3, Mar., 1977, pp. 114-122.
22. Orr, D. M. F., "Lap Splicing of Deformed Reinforcing Bars," Journal of the
American Concrete Institute, Vol. 73, No. 11, Nov., 1976, pp. 622-627.
23. Perry, E. S., and Jundi, N., "Pullout Bond Stress Distribution Under Static
and Dynamic Repeated Loadings," Journal of the American Concrete Institute,
Vol. 66, No. 5, May, 1969, pp. 377-380.
24. Pfister, J. F., and Mattock, A. H., "High Strength Bars at Concrete Rein-
forcement Park 5. Lapped Splices in Concentrically Loaded Columns," Jour-
nal of the PCA Research and Development Laboratories, Vol. 5, No. 2, May, 1963,
pp. 1-40. (Also reprinted as PCA Development Department Bulletin D63.)
25. Popov, E. P., "Mechanical Characteristics and Bond of Reinforcing Steel Un-
der Seismic Conditions," Proceedings, Workshop on Earthquake-Resistant
Reinforced Concrete Building Construction, July, 1977, pp. 658-682.
26. "Reinforcing Bar Splices," Concrete Reinforcing Steel Institute, 3rd ed., Chi-
cago, 111., 1975, 30 pp.
27. Rezansoff, T., Bufkin, M. P., Jirsa, J. O., and Breen, J. E., "The Performance
of Lapped Splices Under Rapid Loading," Research Report 154-2, Center for
Highway Research, The University of Texas at Austin, Jan., 1975, 92 pp.
28. Rice, P. F., "Rebar Splices: Cutting Costs, Avoiding Errors," Civil Engineer-
ing, ASCE, Vol. 47, No. 8, Aug., 1977, pp. 73-74.
29. Somerville, G., and Taylor, H. P. J., "The Influence of Reinforcement De-
tailing on the Strength of Concrete Structures," The Structural Engineer, Vol.
50, No. 1, Jan., 1972, pp. 7-19.
30. Tepfers, R., "A Theory of Bond Applied to Overlapped Tensile Reinforce-
ment Splices for Deformed Bars," Publication 73:2, Division of Concrete
Structures, Chalmers University of Technology, Goteborg, Sweden, 1973, 328
pp.
31. Thompson, M. A., Jirsa, J. A., Breen, J. E., and Meinheit, D. F., "The Be-
havior of Multiple Lap Splices in Wide Sections," Research Report 154-1, Cen-

.857

J. Struct. Eng. 1983.109:843-858.

 ter for Highway Research, The University of Texas at Austin, Jan., 1975, 75
pp.
32. "Uniform Building Code," International Conference of Building Officials,
Whittier, Calif., 1976, pp: 300-406.
33. Untrauer, R. E., and Warren, G. E., "Stress Development of Tension Steel
in Beams," Journal of the American Concrete Institute, Vol. 74, No. 8, Aug.,
1977, pp. 368-372.
Downloaded from ascelibrary.org by New York University on 05/11/15. Copyright ASCE. For personal use only; all rights reserved.

858

J. Struct. Eng. 1983.109:843-858.